Making Contextual Decisions with Low Technical Debt

By Alekh Agarwal, Sarah Bird, Markus Cozowicz, Luong Hoang, J. Langford, Stephen Lee, Jiaji Li, Daniel R. Melamed, Gal Oshri, Oswaldo Ribas and 2 more. 2016

122 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Balanced Linear Contextual Bandits (2018, background): “…2014) in the case of non-parametric function of arm rewards), has demonstrated successful applications (e.g., news article recommendations (Li et al. 2010) or mobile health (Lei, Tewari, and Murphy 2017)), and has proposed system designs to apply these algorithms in practice (Agarwal et al. 2016).”
Estimation Considerations in Contextual Bandits (2017, background): “, news article recommendations [45] or mobile health [42]), and has proposed system designs to apply these algorithms in practice [3].”
Horizon: Facebook's Open Source Applied Reinforcement Learning Platform (2018, background): “Currently, several platforms have been developed that address different parts of this end-to-end applied RL challenge [2, 4, 10, 29], however to our knowledge, no single system offers an end-toend solution.”
Federated Residual Learning (2020, background): “This enables CB learning to leverage advantage of personalization to individual clients while leveraging joint learning across multiple users as in a fully centralized setting, while prior approaches typically rely only on centralized learning (Agarwal et al., 2016).”
A Contextual Bandit Bake-off (2018, methodology): “All of our experiments are based on the online learning system Vowpal Wabbit 1 which has already been successfully used in production systems (Agarwal et al., 2016).”
Lessons from Real-World Reinforcement Learning in a Customer Support Bot (2019, methodology): “The Decision Service is the basis of our system.”
An empirical investigation of the challenges of real-world reinforcement learning (2020, methodology): “2018] and Decision Service [Agarwal et al. 2016] provide software platforms for training, evaluation and deployment of RL agents in real-world systems.”
Practical Evaluation and Optimization of Contextual Bandit Algorithms (2018, methodology): “All our experiments are based on the online learning system Vowpal Wabbit1 which has already been successfully used in production systems (Agarwal et al., 2016).”
Adversarial Attacks on Stochastic Bandits (2018, background): “Finally, we note that our setting is motivated by modern industry-scale applications of contextual bandits, where arm selection, reward signal collection, and policy updates are done in a distributed way [3, 18].”
Bandit Algorithms (2020, methodology): “Microsoft offers a ‘Decision Service’ that uses bandit algorithms to automate decision-making [Agarwal et al., 2016].”
Challenges of real-world reinforcement learning: definitions, benchmarks and analysis (2021, methodology): “2018) and Decision Service (Agarwal et al. 2016) provide software platforms for training, evaluation and deployment of RL agents in realworld systems.”